Some of the recommended links I read in the last 3 months while I was in the parental leave.
June
AI & LLM
- The New Skill in AI is Not Prompting, It's Context Engineering
- How to Fix Your Context | Drew Breunig
- Project Vend: Can Claude run a small shop? (And why does that matter?)
- GitHub Next | Continuous AI
- Context engineering
- LLMs bring new nature of abstraction
- Agentic Coding Recommendations
- Why Your AI Coding Assistant Keeps Doing It Wrong, and How To Fix It
- Writing documentation for AI: best practices
- How OpenElections Uses LLMs
- An Introduction to Google's Approach to AI Agent Security
- The Gentle Singularity - Sam Altman
- How I program with Agents
- Real-world engineering challenges: building Cursor
- Knowledge Management in the Age of AI | Eric Gardner
- The Prompt Engineering Playbook for Programmers
- My AI Skeptic Friends Are All Nuts · The Fly Blog
- The Copilot Delusion
- How much does AI impact development speed? - by Abi Noda
- Expert Generalists
Architecture & Practices
- Journey to 1000 models: Scaling Instagram's recommendation system - Engineering at Meta
- Swift at Apple: migrating the Password Monitoring service from Java
- Migrating 3.7 Million Lines of Flow Code to TypeScript | by Pinterest Engineering | Pinterest Engineering Blog
- 2000 Lines Of Code
July
AI & LLM
- Agent Interaction Guidelines (AIG) – Linear Developers
- Measuring engineering productivity in the AI era
- MCP の認証と認可の現在と未来 | hi120ki
- Exploring RAG based approaches for Text-to-SQL
- Text-to-SQL with GenAI: A Practical Exploration
- LLM Embeddings Explained: A Visual and Intuitive Guide
- Claude Code weekly rate limits | Hacker News
- Enough AI copilots! We need AI HUDs
- Hierarchical Reasoning Model
- Claude Code is a Slot Machine
- Why I'm Betting Against AI Agents in 2025 (Despite Building Them)
- PerfInsights: Detecting Performance Optimization Opportunities in Go Code using Generative AI | Uber Blog
- Measuring the impact of AI on software engineering – with Laura Tacho
- Stop Building AI Tools Backwards | Hazel Weakly
- Coding with LLMs in the summer of 2025 (an update)
- How I keep up with AI progress (and why you must too) - nilenso blog
- Reflections on OpenAI
- Application development without programmers
- How I use Claude Code (+ my best tips)
- Introduction | LLM Inference in Production
- AI productivity gains are being offset by organizational bottlenecks
- AI Agent Benchmarks are Broken - by Daniel Kang
- Grok: searching X for "from:elonmusk (Israel OR Palestine OR Hamas OR Gaza)"
- Claude Code & the evolution of agentic coding — Boris Cherny, Anthropic - YouTube
- Stop Building AI Agents: Use Smarter LLM Workflows
- Introducing the AI Measurement Framework
- I used o3 to profile myself from my saved Pocket links | noperator
- I Shipped a macOS App Built Entirely by Claude Code | Indragie Karunaratne
- Vibe code is legacy code | Val Town Blog
- I Know When You're Vibe Coding | Alex Kondov - Software Engineer
Architecture & Practices
- The Architecture Behind Lovable and Bolt • Beam
- Local First Software Is Easier to Scale
- I tried Servo, the undercover web browser engine made with Rust
- Vanilla JavaScript support for Tailwind Plus - Tailwind CSS
- How I build software quickly
- Identify, solve, verify
- SPA to SSR and Everything in Between
Leadership & Organization
- How to communicate mandates (even if you disagree with them) - LeadDev
- Measuring Developer Productivity: the good, the bad and the data
August
AI & LLM
- 6 Weeks of Claude Code - Puzzmo Blog
- Making a Model Context Protocol server more robust, and much more private
- Deep Agents
- Diff Risk Score: AI-driven risk-aware software development - Engineering at Meta
- Laurence Tratt: LLM Inflation
- Getting Good Results from Claude Code • Chris Dzombak
- Complex Agentic Coding with Copilot: GPT-5 vs Claude 4 Sonnet
- AgentFlayer: When a Jira Ticket Can Steal Your Secrets
- I: Claude Is the Drug, Cursor Is the Dealer - by Logan Zev
- Future AI bills of $100k/yr per dev - by Ewa Szyszka
- uReview: Scalable, Trustworthy GenAI for Code Review at Uber | Uber Blog
- Why LLMs Can't Really Build Software — Zed's Blog
- Best Practices for Building Agentic AI Systems: What Actually Works in Production - UserJot
Architecture & Practices
- QUIC for the kernel [LWN.net]
- Modern Node.js Patterns for 2025
- How we replaced Elasticsearch and MongoDB with Rust and RocksDB
- Linear sent me down a local-first rabbit hole | Bytemash
- Unexpected benefits of going local-first
- Announcing Neki — PlanetScale
- Everything I know about good system design
- Lessons learned from building a sync-engine and reactivity system with SQLite — Antoine's blog
- New Method Is the Fastest Way To Find the Best Routes | Quanta Magazine
- OpenSSH: Post-Quantum Cryptography
- Maintainers of Last Resort
- iPhone DevOps - ultimate edition